Abstract
There has been a growing trend among users of social media platforms to express their emotions using visual content. Visual sentiment analysis is the process of understanding the emotional polarity of images or videos and is still considered a challenging problem in artificial intelligence. Most of the existing models are based on robust machine learning or deep learning techniques. The idea of using deep transfer learning techniques for visual sentiment analysis is fairly new. In this paper, we propose a new approach using data-augmented-transfer learning architecture consisting of a pre-trained VGG16 model that is fine-tuned using SVM with augmented training data. For fine-tuning and evaluation, we initially use two Twitter image datasets. We further validated the proposed model on a third dataset. The commonly used geometric augmentation methods such as rotation, zoom range, width shift, height shift, shear range and horizontal flip were are used. We compare our proposed VGG16-SVM model with 3 other state-of-the-art deep models commonly used for transfer learning and 4 machine learning models (besides SVM) used for fine-tuning. The results show that VGG16-SVM produces the overall best accuracy (94%) and recall (96%) among all transfer learning and machine learning pairs. We also show that our proposed model outperforms all previous studies that use the same dataset.
Similar content being viewed by others
Availability of data and material
No dataset generated in this study.
References
Ahsan U, De Choudhury M, Essa I (2017) Towards using visual attributes to infer image sentiment of social events. IEEE, pp 1372–1379
Ain QT et al (2017) Sentiment analysis using deep learning techniques: a review. Int J Adv Comput Sci Appl 8(6):424
Borth D, Chen T, Ji R, Chang S-F (2013) Sentibank: large-scale ontology and classifiers for detecting sentiment and emotions in visual content. pp 459–460
Borth D, Ji R, Chen T, Breuel T, Chang S-F (2013) Large-scale visual sentiment ontology and detectors using adjective noun pairs. pp 223–232
Cai G, Lv G (2017) Heterogeneous transfer with deep latent correlation for sentiment analysis, vol 2. IEEE, pp 252–256
Cambria E, Livingstone A, Hussain A (2012) In: The hourglass of emotions. Springer, pp 144–157
Campos V, Jou B, Giro-i Nieto X (2017) From pixels to sentiment: fine-tuning CNNs for visual sentiment prediction. Image Vis Comput 65:15–22
Chandrasekaran G, Antoanela N, Andrei G, Monica C, Hemanth J (2022) Visual sentiment analysis using deep learning models with social media data. Appl Sci 12(3):1030
Chen S, Yang J, Feng J, Gu Y (2017) Image sentiment analysis using supervised collective matrix factorization. IEEE, pp 1033–1038
Fan S et al (2017) The role of visual attention in sentiment prediction. pp 217–225
Fengjiao W, Aono M (2018) Visual sentiment prediction by merging hand-craft and CNN features. IEEE, pp 66–71
Giancristofaro GT, Panangadan A (2016) Predicting sentiment toward transportation in social media using visual and textual features. IEEE, pp 2113–2118
Göring S, Brand K, Raake A (2018) Extended features using machine learning techniques for photo liking prediction. IEEE, pp 1–6
Hassan SZ et al (2020) Visual sentiment analysis from disaster images in social media. Preprint at http://arxiv.org/abs/2009.03051
Huang C-C, Wu Y-L, Tang C-Y (2019) Human face sentiment classification using synthetic sentiment images with deep convolutional neural networks. IEEE, pp 1–5
Islam J, Zhang Y (2016) Visual sentiment analysis for social images using transfer learning approach. IEEE, pp 124–130
Ji R, Cao D, Zhou Y, Chen F (2016) Survey of visual sentiment prediction for social media analysis. Front Comp Sci 10(4):602–611
Li Z et al (2021) Visual sentiment analysis based on image caption and adjective–noun–pair description. Soft Computing 1–13
Li Z, Fan Y, Jiang B, Lei T, Liu W (2019) A survey on sentiment analysis and opinion mining for social multimedia. Multimed Tools Appl 78(6):6939–6967
Liu W, Qiu J-L, Zheng W-L, Lu B-L (2019) Multimodal emotion recognition using deep canonical correlation analysis. Preprint at http://arxiv.org/abs/1908.05349
McDuff D, El Kaliouby R, Cohn JF, Picard RW (2014) Predicting ad liking and purchase intent: large-scale analysis of facial responses to ads. IEEE Trans Affect Comput 6(3):223–235
Mittal N, Sharma D, Joshi ML (2018) Image sentiment analysis using deep learning. IEEE, pp 684–687
Poria S et al (2017) Context-dependent sentiment analysis in user-generated videos. pp 873–883
Poria S, Peng H, Hussain A, Howard N, Cambria E (2017) Ensemble application of convolutional neural networks and multiple kernel learning for multimodal sentiment analysis. Neurocomputing 261:217–230
Siersdorfer S, Minack E, Deng F, Hare J (2010) Analyzing and predicting sentiment of images on the social web. pp 715–718
Sun M, Yang J, Wang K, Shen H (2016) Discovering affective regions in deep convolutional neural networks for visual sentiment prediction. IEEE, pp 1–6
Vadicamo L et al (2017) Cross-media learning for image sentiment analysis in the wild. pp 308–317
Wang Y, Li B (2015) Sentiment analysis for social media images. IEEE, pp 1584–1591
Wang C, Yang J, Xie L, Yuan J (2019) Kervolutional neural networks. pp 31–40
Wu L, Zhang H, Deng S, Shi G, Liu X (2021) Discovering sentimental interaction via graph convolutional network for visual sentiment prediction. Appl Sci 11(4):1404
Wu L, Zhang H, Shi G, Deng S (2021) Weakly supervised interaction discovery network for image sentiment analysis. Tech. Rep, EasyChair
Yadav A, Vishwakarma DK (2020) Sentiment analysis using deep learning architectures: a review. Artif Intell Rev 53(6):4335–4385
Yang J et al (2018) Visual sentiment prediction based on automatic discovery of affective regions. IEEE Trans Multimedia 20(9):2513–2525
Yang L, Song Q, Wang Z, Jiang M (2019) Parsing r-CNN for instance-level human analysis. pp 364–373
Yazdavar AH et al (2020) Multimodal mental health analysis in social media. PLoS ONE 15(4):e0226248
You Q, Jin H, Luo J (2017) Visual sentiment analysis by attending on local image regions
You Q, Luo J, Jin H, Yang J (2015) Robust image sentiment analysis using progressively trained and domain transferred deep networks
You Q, Luo J, Jin H, Yang J (2015) Robust image sentiment analysis using progressively trained and domain transferred deep networks
Yuan J, Mcdonough S, You Q, Luo J (2013) Sentribute: image sentiment analysis from a mid-level perspective. pp 1–8
Zhang K, Zhu Y, Zhang W, Zhu Y (2021) Cross-modal image sentiment analysis via deep correlation of textual semantic. Knowl-Based Syst 216:106803
Zisad SN, Chowdhury E, Hossain MS, Islam RU, Andersson K (2021) An integrated deep learning and belief rule-based expert system for visual sentiment analysis under uncertainty. Algorithms 14(7):213
Zisad SN, Chowdhury E, Hossain MS, Islam RU, Andersson K (2021) An integrated deep learning and belief rule-based expert system for visual sentiment analysis under uncertainty. Algorithms 14(7):213
Author information
Authors and Affiliations
Contributions
All author contributed equally.
Corresponding author
Ethics declarations
Compliance with ethical standards
This statement is to certify that the author list is correct. The Authors also confirm that this research has not been published previously and that it is not under consideration for publication elsewhere. On behalf of all Co-Authors, the Corresponding Author shall bear full responsibility for the submission. There is no conflict of interest. This research did not involve any human participants and/or animals.
Conflict of interest
The authors declare that they have no relevant financial or non-financial interests to disclose. There is no personal relationship that could influence the work reported in this paper. No funding was received for conducting this study. The authors have no conflicts of interest to declare that are relevant to the content of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Jiang, Z., Zaheer, W., Wali, A. et al. Visual sentiment analysis using data-augmented deep transfer learning techniques. Multimed Tools Appl 83, 17233–17249 (2024). https://doi.org/10.1007/s11042-023-16262-4
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16262-4